# -*- coding: utf-8 -*-
# Turn off GPU if needed
import os
os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
##Checking if using GPU
import tensorflow as tf
from tensorflow.python.client import device_lib
print(device_lib.list_local_devices())
print("Num GPUs Available: ", len(tf.config.experimental.list_physical_devices('GPU')))
tf.test.is_gpu_available(
cuda_only=False, min_cuda_compute_capability=None
)
tf.config.experimental.list_physical_devices('GPU')
"""# SET UP
"""
!pip install tensorflow-gpu==2.3.0
!pip install Keras==2.4.3
import keras
import tensorflow as tf
print(keras.__version__,tf.__version__)
###############################
####### Start from here #######
###############################
import keras
import tensorflow as tf
import numpy as np
import warnings
import argparse
warnings.filterwarnings("ignore")
from keras.models import load_model
from keras.optimizers import Adam
import h5py
import IPython
!pip install -q -U keras-tuner
import kerastuner as kt
# Commented out IPython magic to ensure Python compatibility.
# %load_ext autoreload
# %autoreload 2
# %matplotlib inline
"""# Load Data & Preprocessing
"""
X_train= np.load('/data/Xtrain.npy')
y_train=np.load('/data/ytrain.npy')
X_test=np.load('/data/Xtest.npy')
y_test=np.load('/data/ytest.npy')
#X_train=X_train[:,0,:,:]
#X_test=X_test[:,0,:,:]
import numpy as np
def preprocess(X):
m=4096-X.shape[2]
y=np.pad(X_train,[(0,0),(0,m),(0,0)],mode='constant', constant_values=0)
return y
X_train=preprocess(X_train)
X_test=preprocess(X_test)
print(X_train.shape,X_test.shape)
"""## Residual Net Model
"""
from keras.layers import (Input, Conv1D, MaxPooling1D, Dropout,
BatchNormalization, Activation, Add,
Flatten, Dense)
from keras.models import Model
import numpy as np
class ResidualUnit(object):
"""Residual unit block (unidimensional).
Parameters
----------
n_samples_out: int
Number of output samples.
n_filters_out: int
Number of output filters.
kernel_initializer: str, optional
Initializer for the weights matrices. See Keras initializers. By default it uses
'he_normal'.
dropout_rate: float [0, 1), optional
Dropout rate used in all Dropout layers. Default is 0.8
kernel_size: int, optional
Kernel size for convolutional layers. Default is 17.
preactivation: bool, optional
When preactivation is true use full preactivation architecture proposed
in [1]. Otherwise, use architecture proposed in the original ResNet
paper [2]. By default it is true.
postactivation_bn: bool, optional
Defines if you use batch normalization before or after the activation layer (there
seems to be some advantages in some cases:
https://github.com/ducha-aiki/caffenet-benchmark/blob/master/batchnorm.md).
If true, the batch normalization is used before the activation
function, otherwise the activation comes first, as it is usually done.
By default it is false.
activation_function: string, optional
Keras activation function to be used. By default 'relu'.
References
----------
.. [1] K. He, X. Zhang, S. Ren, and J. Sun, "Identity Mappings in Deep Residual Networks,"
arXiv:1603.05027 [cs], Mar. 2016. https://arxiv.org/pdf/1603.05027.pdf.
.. [2] K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning for Image Recognition," in 2016 IEEE Conference
on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770-778. https://arxiv.org/pdf/1512.03385.pdf
"""
def __init__(self, n_samples_out, n_filters_out, kernel_initializer='he_normal',
dropout_rate=0.5, kernel_size=17, preactivation=True,
postactivation_bn=False, activation_function='relu'):
self.n_samples_out = n_samples_out
self.n_filters_out = n_filters_out
self.kernel_initializer = kernel_initializer
self.dropout_rate = dropout_rate
self.kernel_size = kernel_size
self.preactivation = preactivation
self.postactivation_bn = postactivation_bn
self.activation_function = activation_function
def _skip_connection(self, y, downsample, n_filters_in):
"""Implement skip connection."""
# Deal with downsampling
if downsample > 1:
y = MaxPooling1D(downsample, strides=downsample, padding='same')(y)
elif downsample == 1:
y = y
else:
raise ValueError("Number of samples should always decrease.")
# Deal with n_filters dimension increase
if n_filters_in != self.n_filters_out:
# This is one of the two alternatives presented in ResNet paper
# Other option is to just fill the matrix with zeros.
y = Conv1D(self.n_filters_out, 1, padding='same',
use_bias=False, kernel_initializer=self.kernel_initializer)(y)
return y
def _batch_norm_plus_activation(self, x):
if self.postactivation_bn:
x = Activation(self.activation_function)(x)
x = BatchNormalization(center=False, scale=False)(x)
else:
x = BatchNormalization()(x)
x = Activation(self.activation_function)(x)
return x
def on_epoch_end(self, epoch, logs=None):
print('###########',Keras.eval(self.model.optimizer.lr))
def __call__(self, inputs):
"""Residual unit."""
x, y = inputs
#n_samples_in = y.shape[1].value
n_samples_in = y.shape[1]
downsample = n_samples_in // self.n_samples_out
#n_filters_in = y.shape[2].value
n_filters_in = y.shape[2]
y = self._skip_connection(y, downsample, n_filters_in)
# 1st layer
x = Conv1D(self.n_filters_out, self.kernel_size, padding='same',
use_bias=False, kernel_initializer=self.kernel_initializer)(x)
x = self._batch_norm_plus_activation(x)
if self.dropout_rate > 0:
x = Dropout(self.dropout_rate)(x)
# 2nd layer
x = Conv1D(self.n_filters_out, self.kernel_size, strides=downsample,
padding='same', use_bias=False,
kernel_initializer=self.kernel_initializer)(x)
if self.preactivation:
x = Add()([x, y]) # Sum skip connection and main connection
y = x
x = self._batch_norm_plus_activation(x)
if self.dropout_rate > 0:
x = Dropout(self.dropout_rate)(x)
else:
x = BatchNormalization()(x)
x = Add()([x, y]) # Sum skip connection and main connection
x = Activation(self.activation_function)(x)
if self.dropout_rate > 0:
x = Dropout(self.dropout_rate)(x)
y = x
return [x, y]
# ----- Model ----- #
kernel_size = 8
kernel_initializer = 'he_normal'
signal = Input(shape=(4096, 12), dtype=np.float32, name='signal')
age_range = Input(shape=(6,), dtype=np.float32, name='age_range')
is_male = Input(shape=(1,), dtype=np.float32, name='is_male')
x = signal
x = Conv1D(64, kernel_size, padding='same', use_bias=False,
kernel_initializer=kernel_initializer)(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x, y = ResidualUnit(1024, 128, kernel_size=kernel_size,
kernel_initializer=kernel_initializer)([x, x])
x, y = ResidualUnit(256, 196, kernel_size=kernel_size,
kernel_initializer=kernel_initializer)([x, y])
x, y = ResidualUnit(64, 256, kernel_size=kernel_size,
kernel_initializer=kernel_initializer)([x, y])
x, _ = ResidualUnit(16, 320, kernel_size=kernel_size,
kernel_initializer=kernel_initializer)([x, y])
x = Flatten()(x)
#diagn = Dense(6, activation='sigmoid')(x)
diagn = Dense(1,activation='sigmoid')(x)
model = Model(signal, diagn)
model.save_weights('model.h5')
model.summary()
import tensorflow as tf
bce_logits = tf.keras.losses.BinaryCrossentropy(from_logits=True, name='binary_crossentropy')
lr = 0.01
opt = tf.keras.optimizers.Adam(lr,epsilon=0.1)
loss = bce_logits
lr_metric = opt.lr
#model.compile(loss='binary_crossentropy', optimizer=optimizer, metrics=['acc', lr_metric])
import keras
from datetime import datetime
#from sklearn.metrics import confusion_matrix, f1_score,
def scheduler(epoch, lr):
if epoch <= 7:
tf.summary.scalar('learning rate', data=lr, step=epoch)
return lr
else:
tf.summary.scalar('learning rate', data=lr * 0.9, step=epoch)
return lr * 0.9
callback = tf.keras.callbacks.LearningRateScheduler(scheduler)
#es = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=4)
logdir = "logs/scalars/" + datetime.now().strftime("%Y%m%d-%H%M%S")
tensorboard_callback = keras.callbacks.TensorBoard(log_dir=logdir)
bce_logits = tf.keras.losses.BinaryCrossentropy(from_logits=True, name='binary_crossentropy')
lr = 0.005
opt = tf.keras.optimizers.Adam(lr,epsilon=0.1)
loss = bce_logits
lr_metric = opt.lr
model.compile(optimizer=opt,
loss='binary_crossentropy',
metrics=[tf.keras.metrics.BinaryAccuracy(), tf.keras.metrics.AUC(),
tf.keras.metrics.Precision(), tf.keras.metrics.Recall()])
model.fit(X_train,
y_train,
batch_size=64,
epochs=50,
validation_data=(X_test, y_test),
callbacks=[callback,tensorboard_callback])
# Commented out IPython magic to ensure Python compatibility.
# %load_ext tensorboard
# %tensorboard --logdir logs
# Commented out IPython magic to ensure Python compatibility.
# %tensorboard --logdir logs
"""# Hyperparameter Tuning
"""
from kerastuner import HyperModel
from keras import Sequential
from keras.layers import (Input, Conv1D, MaxPooling1D, Dropout,
BatchNormalization, Activation, Add,
Flatten, Dense)
from keras.models import Model
import numpy as np
class ECGhyper(HyperModel):
def __init__(self, input_shape):
self.input_shape = input_shape
def build(self, hp):
'''
kernel_initializer = 'he_normal'
model = Sequential()
model.add(
Conv1D(64,
kernel_size=hp.Choice('kernel_size', [8,16,24,38], default=8),
input_shape=input_shape
)
)
model.add(BatchNormalization())
model.add(Activation('relu'))
model.add(
ResidualUnit(1024, 128,
kernel_size=hp.Choice('kernel_size', [8,16,24,38], default=8),
kernel_initializer=kernel_initializer
)
)
model.add(
ResidualUnit(256, 196, kernel_size=hp.Choice('kernel_size', [8,16,24,38], default=8),
kernel_initializer=kernel_initializer))
model.add(
ResidualUnit(64, 256, kernel_size=hp.Choice('kernel_size', [8,16,24,38], default=8),
kernel_initializer=kernel_initializer))
model.add(
ResidualUnit(16, 320, kernel_size=hp.Choice('kernel_size', [8,16,24,38], default=8),
kernel_initializer=kernel_initializer))
model.add(Flatten())
model.add(Dense(1,activation='sigmoid'))
'''
#kernel_size = 16
kernel_size=hp.Choice('kernel_size', [8,16,24,38], default=8)
kernel_initializer = 'he_normal'
signal = Input(shape=(4096, 12), dtype=np.float32, name='signal')
x = signal
x = Conv1D(64,
kernel_size=kernel_size,
input_shape=input_shape,kernel_initializer=kernel_initializer
)(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x, y = ResidualUnit(1024, 128,
kernel_size=kernel_size,
kernel_initializer=kernel_initializer
)([x, x])
x, y = ResidualUnit(256, 196, kernel_size=kernel_size,
kernel_initializer=kernel_initializer)([x, y])
x, y = ResidualUnit(64, 256, kernel_size=kernel_size,
kernel_initializer=kernel_initializer)([x, y])
x, _ = ResidualUnit(kernel_size, 320, kernel_size=kernel_size,
kernel_initializer=kernel_initializer)([x, y])
x = Flatten()(x)
#diagn = Dense(6, activation='sigmoid')(x)
diagn = Dense(1,activation='sigmoid')(x)
model = Model(signal, diagn)
lr = hp.Choice('lr', values = [1e-2, 1e-3, 1e-4])
#bce_logits = tf.keras.losses.BinaryCrossentropy(from_logits=True, name='binary_crossentropy')
opt = tf.keras.optimizers.Adam(lr,epsilon=0.1)
#loss = bce_logits
#lr_metric = opt.lr
model.compile(optimizer=opt,
loss='binary_crossentropy',
metrics=[tf.keras.metrics.BinaryAccuracy(), 'AUC',
'Precision', 'Recall'])
return model
input_shape=(4096, 12)
hypermodel=ECGhyper(input_shape)
from kerastuner import RandomSearch, Objective
tuner_rs = RandomSearch(
hypermodel,
objective=Objective("val_loss", direction="max"),
seed=42,
max_trials=10,
executions_per_trial=2)
import keras
from datetime import datetime
#from sklearn.metrics import confusion_matrix, f1_score,
def scheduler(epoch, lr):
if epoch <= 7:
tf.summary.scalar('learning rate', data=lr, step=epoch)
return lr
else:
tf.summary.scalar('learning rate', data=lr * 0.9, step=epoch)
return lr * 0.9
callback = tf.keras.callbacks.LearningRateScheduler(scheduler)
#es = EarlyStopping(monitor='val_loss', mode='min', verbose=1, patience=4)
logdir = "logs/scalars/" + datetime.now().strftime("%Y%m%d-%H%M%S")
tensorboard_callback = keras.callbacks.TensorBoard(log_dir=logdir)
tuner_rs.search(X_train, y_train, epochs=10, validation_split=0.2,callbacks=[callback,tensorboard_callback])
best_model = tuner_rs.get_best_models(num_models=1)[0]
#loss, mse = best_model.evaluate(x_test_scaled, y_test)
best_model.fit(X_train,
y_train,
batch_size=64,
epochs=30,
validation_data=(X_test, y_test),
callbacks=[callback,tensorboard_callback])
# Commented out IPython magic to ensure Python compatibility.
# %load_ext tensorboard
# %tensorboard --logdir logs
model.summary()
model.save_weights('model.h5')
#####RESET MODEL######
model.load_weights('model.h5')
Datasets
Models
